In this application, we seek to demonstrate a new subtractive method to remove abundant proteins from complex human fluid samples. The ultimate goal is to develop a suite of methods or tools to facilitate the de novo discovery of a complete set of low-abundance potential protein biomarkers. More targeted studies then can be performed to result in a protein diagnostic panel. A particular interest is in the discovery of potential biomarkers for cancers; in the past we have focused on lung cancer markers. We seek to demonstrate the selective and specific removal of human serum albumin and immunoglobulin G isoforms and multimers from human plasma. The albumins IgG comprise about two thirds of the total protein in human serum or plasma, and cause dynamic range issues for both tandem mass spectrometry and difference gel electrophoresis approaches for biomarker discovery. Subsequent work builds on the base established here, and would extend to work to remove three orders of magnitude of abundant proteins, a 50 time improvement over existing state-of-the-art methods. One key feature is that the conditions are chosen to minimize protein-protein interactions, which compromise the performance of existing methods. When this process is combined with ever-improving LC/LC-MS/MS and high dynamic range DIGE imagers, nearly the entire ten order of magnitude dynamic range can be explored, facilitating the search for new potential protein biomarkers.